Covalent attachment of the hydrophilic polymer poly(ethylene glycol), abbreviated PEG, also known as poly(ethylene oxide), abbreviated PEO, to molecules and surfaces is of considerable utility in biotechnology and medicine. PEG is a polymer having the beneficial properties of solubility in water and in many organic solvents, lack of toxicity, and lack of immunogenicity. One use of PEG is to covalently attach the polymer to water-insoluble molecules to improve the solubility of the resulting PEG-molecule conjugate. For example, it has been shown that the water-insoluble drug paclitaxel, when coupled to PEG, becomes water-soluble. Greenwald, et al., J. Org. Chem., 60:331–336 (1995). PEG has also been used increasingly in the modification of polypeptide and protein therapeutics.
The use of polypeptides, including proteins, for therapeutic applications has expanded in recent years mainly due to both improved methods for recombinant expression of human polypeptides from various expression systems and improved methods of delivery in vivo. Many of the drawbacks associated with polypeptide therapeutics, including short circulating half-life, immunogenicity and proteolytic degradation, have been improved by various approaches including gene therapy, epitope mutations by directed or shuffling mutagenesis, shielding of the epitope regions by natural or synthetic polymers, fusion proteins, and incorporation of the polypeptide into drug delivery vehicles for protection and slow release.
Polymer modification of proteins, such as covalent attachment of poly(ethylene glycol), has gained popularity as a method to improve the pharmacological and biological properties of therapeutically useful proteins. For example, certain poly(ethylene glycol) conjugated proteins have been shown to have significantly enhanced plasma half-life, reduced antigenicity and immunogenicity, increased solubility and decreased proteolytic degradation when compared to their non-pegylated counterparts. Factors that affect the foregoing properties are numerous and include the nature of the protein itself, the number of poly(ethylene glycol) or other polymer chains attached to the protein, the molecular weight and structure of the polymer chains attached to the protein, the chemistries (i.e., the particular linkers) used to attach the polymer to the protein, and the location of the polymer modified-sites on the protein.
To couple PEG to a molecule, such as a protein, it is often necessary to “activate” the PEG by preparing a derivative of the PEG having a functional group at a terminus thereof. The functional group is chosen based on the type of available reactive group on the molecule that will be coupled to the PEG. For example, the functional group could be chosen to react with an amino group on a protein in order to form a PEG-protein conjugate.
A variety of methods have been developed to non-specifically or randomly attach poly(ethylene glycol) to proteins. Most commonly, electrophilically-activated poly(ethylene glycol) is reacted with nucleophilic side chains found of proteins. Attaching an activated poly(ethylene glycol) to the α-amine and ε-amine groups found on lysine residues and at the N-terminus results in a mixture of conjugate products as described in U.S. Pat. No. 6,057,292. For example, the conjugate may consist of a population of conjugated proteins having varying numbers of poly(ethylene glycol) molecules attached to the protein molecule (“PEGmers”), ranging from zero to the number of α- and ε-amine groups in the protein. Often, random pegylation approaches are undesirable, due to variations in the ratios of PEG-mer products produced, and the desire, in certain cases, for a single, discrete PEG-protein conjugate product. For a protein molecule that has been singly modified by employing a non-site specific pegylation methodology, the poly(ethylene glycol) moiety may be attached at any one of a number of different amine sites. Additionally, this type of non-specific PEGylation can result in partial or complete loss of the therapeutic utility of the conjugated protein, particularly for conjugates having more than one PEG attached to the protein.
Several methods for site-directed or selective attachment of PEG have been described. For example, WO 99/45026 suggests chemical modification of a N-terminal serine residue to form an aldehyde functionality suitable for reaction with a polymer terminated with a hydrazide or semicarbazide functionality. U.S. Pat. Nos. 5,824,784 and 5,985,265 suggest reacting a polymer bearing a carbonyl group with the amino terminus of a protein under reducing alkylation conditions and at a pH that promotes selective attack at the N-terminus. WO 99/03887 and U.S. Pat. Nos. 5,206,344 and 5,766,897 relate to the site-directed PEGylation of cysteine residues that have been engineered into the amino acid sequence of proteins (cysteine-added variants). While these methods offer some advantages over non-specific attachment, there is a continuing unmet need for improved methods and reagents for providing site-specific polymer-conjugated proteins that do not require chemical modification of the polypeptide or careful control of certain reaction conditions, such as pH. Additionally, due to the high desirability for modifying a protein at its reactive amino-functionalities, there is a need for improved polymer reagents that react selectively with a specific protein amino group, such as the N-terminal amino group, for preparing protein-polymer conjugates that are not a mixture of PEG-polymer PEGmers but rather have PEG attached to a single, identified site on the protein.